First thanks Andy.
Am 06.01.23 um 02:21 schrieb andy pugh:
1. "|| beta > 1e10"
There was a small error in the function biarc in
file: src/emc/task/emccanon.cc.
I added the condition "beta > 1e10" because without
the file "Ruota_dentata #1=1.ngc" is not working.
The value of 1E10 is only guessed.
There are pre-defined "Fuzz" values available, maybe one of those
would be appropriate?
https://github.com/LinuxCNC/linuxcnc/blob/2.8/src/libnml/posemath/posemath.h#L709
I found values with negative exponent, but not with positive.
How to proceed?
- 1/CART_FUZZ (#define CART_FUZZ (1.0e-8))
- a new constant in posemath.h
- a define in emccanon.c
My debug code is still there and I fear to violate
your coding style a lot.
https://linuxcnc.org/docs/devel/html/code/style-guide.html
I suggest to use indent.
It's probably better to match the surrounding code than to match the
coding style.
I will do this. It takes some time.
I copy the changed code to a new file, run
indent -kr -i8
as suggested here: https://slurm.schedmd.com/coding_style.pdf
and copy it back.
3. Parameter #1
To select the methods from ECME14.pdf NICU, NICL and NICC
see: http://wiki.linuxcnc.org/cgi-bin/wiki.pl?NURBS
#1 in front of G6.2 is used.
See for example: /nc_files/nurbs/G6/Curva_Chitarra #1=1.ngc
Excerpt:
#1=1
G6.2 X59.171 Y23.693 R46.5833333333333 K0 P5 F290
Do you agree with this?
No, I can't say that I like that particularly. Why not use an extra
(optional) letter in the G6.2 command?
G6.2 uses now K, P, R.
The G code interpreter is a black box for me.
Suggestion Q?
5: Interp::nurbs_reset_global_variables();
There are static variables in
/src/emc/rs274ngc/interp_convert.cc line 142.
" setup_pointer settings) //!< pointer to machine settings" ?
I think that those settings are meant to be persistent. What problem
does it cause?
Persistent no. They are used only for calculations
between start of G6.2 and the last line with G6.2 Kx.
See the examples:
https://github.com/jjjjjf/linuxcnc/tree/nurbs_G5_G6_xyz/nc_files/nurbs/G6
If in a g code file a additional nurbs curve with 6.2 starts,
they habe to be reset.
6. G6.3
This is not used. Please check _gees are correct.
Is Gx is the right modal group?
/* 60 */ -1,-1,-1,-1,-1,-1,-1,-1,-1,-1,15,-1,-1,-1,-1,-1,-1,-1,-1,-1,
Looks like no G6 codes are valid?
Please look here:
https://github.com/jjjjjf/linuxcnc/blob/nurbs_G5_G6_xyz/src/emc/rs274ngc/interp_array.cc
/* 60 */ 1, 1, 1, 0,-1,-1,-1,-1,-1,-1,15,-1,-1,-1,-1,-1,-1,-1,-1,-1, // jf
added G6
I copied this from Lo Valvo.
In the code G6.3 is not used.
is this correct?
BTW
Because the nurbs code results in straight lines
and arcs it must be possible to add
cutter radius compensation.
Joachim
_______________________________________________
Emc-developers mailing list
Emc-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-developers
